TPAC - Technology Policy and Assessment Center at the Georgia Institute of Technology

1999 Indicators of Technology-Based Competitiveness of 33 Nations

Summary Report

Report to the Science Indicators Unit, Science Resources Studies Division, National Science Foundation under Purchase Order B04841X-00-0

March 3, 2000

To view in pdf format click here

Background "High Technology Indicators" (HTI) development work at Georgia Tech has proceeded in phases. The first phase generated indicators of national capabilities to produce high technology products suitable for export. We developed a conceptual model of the processes by which industrializing nations gain access to external technology and technical information, absorb that technology/information effectively, and institutionalize a science-based development and manufacturing capability. The model includes four "input" or leading indicators of a nation's future capacity (roughly 15-year time horizon) to compete in international markets in high technology products. It derives three "output" indicators of a nation's current international competitiveness. The research team compiled data for 20 countries representing a range of regions and extent of industrialization to prepare the initial HTI round for 1987.

The second phase used data on an expanded set of 28 countries to examine in detail the indicators' reliability and validity. J.D. Roessner, A.L. Porter, and H. Xu ["National Capacities to Absorb and Institutionalize External Science and Technology," Technology Analysis & Strategic Management, Vol. 4, No. 2 (1992)] summarized the conceptual development and the 1990 HTI.

The third phase of HTI work culminated in HTI 1993, albeit work extended into 1995. We refined formulae for the seven indicators developed in the previous phases, applied them empirically, tested the value of the indicators for policy and scholarly purposes, and assessed the processes of data collection and verification. The results of this phase are incorporated in Figures 9-15 of this report and were discussed in two articles:

HTI were prepared again for 1996, following closely the format developed for 1993. Country coverage is discussed in the next section. Results are included in Figures 9-15 of this report.

Discussion of the 1993 HTI appeared in Science & Engineering Indicators-1993 (National Science Board, Washington, DC, U.S. Government Printing Office, NSF 93-1, p. 186-189); of the 1996 HTI, in Science & Engineering Indicators-1998 (p. 6-33 6-37, Appendix table 6-23). HTI-1993 also contributed to a special report of NSF’s Science Resources Studies Division, Asia’s New High-Tech Competitors (NSF 95-309, 1995).

In sum, HTI have been prepared at 3-year intervals since 1987. Beginning with 1993, the indicators stabilized to permit time series "indicator" comparisons. This report summarizes findings from the 1993, 1996, and 1999 HTI.

Country Coverage Country coverage evolves gradually since 1993. The 1996 HTI adds Poland, Venezuela, and South Africa, but drops Hong Kong, so the total number of countries comes to 30. [Regularization of statistics for China and Hong Kong remains problematic even for 1999, particularly in sorting out exports.] For 1999, HTI adds Ireland, Israel, and the Czech Republic, yielding a total of 33 countries. As in previous reports, the countries are clustered in charts as follows:

Socioeconomic Infrastructure (SE): The social and economic institutions that support and maintain the physical, human, organizational, and economic resources essential to the functioning of a modern, technology-based industrial nation.

Productive Capacity (PC): The physical and human resources devoted to manufacturing products, and the efficiency with which those resources are used.

Technological Standing (TS): An indicator of a country's recent overall success in exporting high technology products.

Technological Emphasis (TE): An indicator of a country's relative emphasis on high technology products in its overall export product mix.

Rate of Technological Change (RTC): An indicator of how rapidly a country is improving its high technology export performance.

The emphasis on electronics reflects our assumption that this has been a vital contributor to much high technology development in recent years. We recognize changing realities that likely warrant broadening the definition of "high tech" in future HTI formulations (see the final paragraph).

Expert Panel
Expert opinion data were obtained from a survey of the International Technology Indicators Panel during summer and fall, 1999. Surveys were largely completed through a website interface [at http://tpac.gatech.edu/], augmented by e-mail, fax, and mail contacts. The resulting group of 303 experts (up from 207 in 1996) collectively provided 336 responses (up from 265). The average number of responses per country was 10.2, ranging from 6 to 22. Only Ireland had 6 responses; 6 others had 7 responses; 8 or more experts assessed the other 26 countries.

Beginning in the late 1980’s, we have worked together with our National Science Foundation colleagues to invite selected professionals to serve on the International Technology Indicators Panel. Our criteria include direct knowledge of the country and of the bases for technological competitiveness. Prototypical experts include embassy science attaches, faculty members, and industrial professionals. Attendees at international conferences and participants in journal advising and publishing relating to technology analysis, forecasting, management, and so forth are good candidates for the Panel. We seek balance among multiple perspectives, and between indigenous and external "watchers." Over time turnover in membership is heavy -- only 24% of the current respondents also participated in 1996. We invite various persons who appear to meet these criteria, but ultimately self-selection comes into play. The respondents indicate their familiarity on a self-report scale item. Due to the general nature of our criteria, we are cautious in interpreting the sector-specific item responses on current and 15-year future prospects.

Results: Input Indicators
We present results for the 33 countries using the previously mentioned groupings. Table 1 consolidates indicator information for the four input indicators and three output indicators, for each of 1993, 1996, and 1999. Figures 1-15 provide a series of bar charts breaking out the indicator data. [NOTE: As discussed in "Country Coverage," we have added 6 countries over this period; 1993 and 1996 values have been constructed for those countries, as feasible, but are missing in several instances.]

National Orientation (NO) indicates a country's commitment to technology-based development along a number of dimensions: government policy, political stability, entrepreneurial spirit, and acceptance of the idea that development should be technology-based. Figure 1 (1999 data) and Figure 9 (1993-96-99 data) present NO. The behavior of NO is quite regular. Recall its formulation it is comprised of weighted responses to 4 expert opinion items (5-point scale responses, so these do not result in extreme differences) plus a scaled investment risk assessment value.

Using the "Big Three" as a benchmark, note how many countries exhibit comparable NO levels commitment to high tech development is widespread. Note also that this includes Eastern Europe.

Two of our newcomers, Ireland and Israel, lead in NO, along with Singapore and Taiwan. Ireland is the only country in 1999 to receive the highest 5-year investment risk rating from Political Risk Services. Singapore and Taiwan have been consistently high on NO in 1993, 1996, and 1999.

Notable declines over the 1993-99 time period (e.g., drops of 10 or more points) show for Sweden, Malaysia, and Thailand. Sweden dropped on each of the five component measures from 1993 to 1999. Declines for Malaysia and Thailand come from drops in estimated investment risk and entrepreneurial spirit. Recall that HTI are relative indicators. Hence, a "decline" on NO or another indicator does not imply an actual drop, just that competing countries in the HTI have advanced faster.

NO Increases of 10 or more points appear for Spain, Canada, Australia, New Zealand, Russia, The Philippines (downturn from 1996 to 1999), and India. Of the component measures that make up NO, increases for Spain, Canada, and Australia were driven mainly by change of social acceptance of technology (Q. 3). NO increases for New Zealand and The Philippines were due mainly to increases in the investment risk index (on our 100-point S scaling, New Zealand increased 18 points and The Philippines 55 points over this 6-year period). Russia showed moderate increases on each of the 5 NO component measures.

Socioeconomic Infrastructure (SE) indicates the strength of each nation's educational system, mobility of capital, and encouragement of foreign investment. Figures 2 and 10 display SE. Strong socioeconomic infrastructure is not restricted to the OECD nations. The three Tigers (Singapore, South Korea, and Taiwan) are striking in their parity with the heavily industrialized nations. Ireland and Israel, as with NO, stack up well on SE. While Singapore, Taiwan, and Ireland lag a bit on tertiary education, their strong SE is not unduly driven by any single component.

Along with the U.S., Canada and Australia lead on SE and show sizable increases from 1993 to 1999. Canada shows especially high on tertiary education and encouragement of foreign-owned firms; Australia, on secondary education and encouragement of foreign-owned firms.

The Eastern and Western European nations display generally increasing SE. The Latin American countries show relative declines. For instance, each of Mexico, Brazil, and Argentina show increases on both secondary and tertiary education participation, yet their Harbison-Myers Index S-scores decline because other countries increased more.

SE interpretation demands some attention to its composition. "HMHS" is the widely used Harbison-Myers Human Skills Index. It includes percentage in secondary and tertiary education. Values range to 148% for Australia for secondary education, reflecting UNESCO data categorization difficulties. However, as with our other indicators, each component is separately normalized to reduce such artifacts. So, the resulting scores benchmark against Canada (S score of 100 = highest of the 33 country set). Singapore’s SE score of 72 equals that of Germany, despite Singapore’s relatively weak HMHS S-score of 48, reflecting its compensating high scores on the two expert opinion items that round out SE.

Technological infrastructure (TI) captures the strength and contributions of a nation's scientific and engineering manpower, its electronic data processing purchases, the relationship of its R&D to industrial application, and its ability to make effective use of technical knowledge. The composition of this indicator includes four expert opinion items plus a measure of purchases of electronic data processing (EDP) equipment, and number of scientists and engineers in R&D. Even on our S-score basis, the U.S. swamps all others on

EDP purchases (U.S. scores 100, trailed by Japan at 39 and Germany at 15, down to Indonesia at 0.3). Numbers of scientists and engineers also vary extremely. Rescaled, the U.S. scores 100, followed by Japan (73) and Russia (67), on down to Malaysia (0.2).

There continues to be much greater variation in TI among the 33 nations than was the case for either NO or SE (Figures 3 and 11). The U.S. stands out, increasing its lead over this time period. Its S-scores range from 91-100 on each component.

The Asian Cubs (particularly Malaysia, Thailand, Indonesia, and the Philippines not China and India) lag, as do the Latin American countries, excepting Brazil. As detailed, the EDP and "number of Scientists & Engineers" components of TI exert heavy influence compared to the four expert opinion items in this patterning. For instance, the countries just noted as relatively low on TI show S-scored on EDP purchases and number of Scientists & Engineers in the range of 0.3 to 2.4, far below their S-scores for the questionnaire items.

Few of the 33 countries show much change in TI from 1993 to 1999. Shifts of 10 or more points appear only for India -- an increase. India shows increases on every TI component.

Productive Capacity (PC) concerns capabilities to manufacture technology-intensive products. It combines the value of electronics production with three survey items related to manufacturing and managerial capabilities to measure the amount and efficiency of resources available. Electronics production values exert considerable influence as they range widely. The U.S. scales at 100 on electronics production followed by Japan at 61, with a marked drop to China (19), Germany (15), UK and South Korea (13), and Singapore (12).

Productive Capacity clearly separates the U.S. and Japan from the rest of the countries in our sample (Figures 4 and 12) with the U.S. usurping top position from Japan. The relative distancing of the U.S. from Japan since 1996 is attributable to shifts in electronics production (changes on the three expert opinion items are small). While most nations showed modest changes in absolute value of electronics production ($Billion), the U.S. increased by $71B and Japan declined by $44B from 1996 to 1999. The U.S. position is so strong that even China’s remarkable doubling of electronics production from $33B to $65B(third in the world behind the U.S. at $345B and Japan at $212B) increases its S-score on this measure only from 12 to 19.

France has overtaken Germany on this indicator, forming the next tier. The French increase is driven by gains on perceived availability of manufacturing labor and industrial management capabilities.

At the next level, the Asian Tigers rate on a par with the remaining OECD countries on PC, aided by their strong electronics production.

Over time, Japan shows a relative decline; Australia, China, and India show notable gains since 1993. China’s is mostly attributable to the increased electronics production already noted. Conversely, gains by Australia and India are attributable completely to the survey items (perceived gains in production skills). Eastern Europe, in general, shows upward tendencies.

Overall: The pattern of the four "leading" indicators shows more of a mixed picture than was the case in the 1993 study. In the earlier study, the (then Four) Tigers had emerged as obvious challengers to the developed nations of the West, with several of the "Asian Cubs" (notably Malaysia, the Philippines, and Indonesia) following not far behind. The 1999 results show considerable consistency but also considerable differences. Regionally, the former Eastern Bloc nations demonstrate largely positive gains on all four indicators from 1993-99. The Latin American countries show negative tendencies on all four indicators.

Results: Output Indicators

High Tech Standing (TS) measures current high tech production and export standing (Figures 5 and 13). TS incorporates three components an expert opinion item (rating technology-intensive production), overall high tech exports, and the value of electronics exports. As noted for the input indicators, the skewed distributions on statistical components exert strong influence on the resulting indicator, even though each component is scaled separately for the 33-country set (S-scores). This effect appears for TS, with the U.S. the benchmark (score of 100) for both overall high tech exports and electronics exports. On overall high tech exports, the U.S. is followed by Japan (59) and Germany (54); for electronics, the U.S. is followed by Japan (98) and Singapore (61).

As in 1993, Japan and the U.S. remain well ahead of all others in high tech competitiveness, however the U.S. has forged into a notable lead over Japan (Figures 5 and 13). The U.S. advance traces back to marked gains on overall high tech exports and on electronics exports; strangely, the U.S. has slipped a bit on the expert opinion measure relative to Japan.

Germany is considerably closer to the other leading nations (UK, Singapore, France) than to the U.S. and Japan on TS. The "Big Three" are more truly "The Big Two" now. This distancing is not due to any decline in Germany, but rather to the remarkable gains by the U.S.

The elevation of Singapore’s position since 1993 is remarkable particularly in that two of the three components that make up TS are absolute (not per capita or otherwise normalized) measures of electronics and high tech exports. Its high tech exports are about 97% electronics, so our inclusion of electronics export as a separate component of TS certainly benefits Singapore. No matter Singapore’s $70B in high tech exports ranks sixth in the world on an absolute basis (the 1999 TS draws upon 1997 export data, the most recent available).

Other nations spread out considerably on the TS measure, generally changing modestly since 1993. An argument could be made that the TS component, X97, high tech exports for 1997 (the most recent year available), is the ‘real’ output target. The disparities in high tech exporting are huge:

"Top 10" High Tech Exporters for 1997

So, the range in high tech exporting in this elite group of 33 nations is from less than $1 billion to $258 billion! [For details on what X97 includes, see the Appendix.]

Big gainers on TS over this time period are Canada, Singapore, and China. China’s move is not at all based on the questionnaire item; it reflects the country’s enormous increase in electronics (and overall high tech) exporting.

The Russian pattern is anomalous (low number of respondents in 1993 and 1996 may contribute to instability here as they spurted in 1996, then fell back in 1999).

Two of the just-added countries show strongly Ireland and Israel. Both are bolstered by highly favorable expert opinion responses, but particularly for Ireland, high tech export data confirm strength.

Technological Emphasis (TE) indicates the degree to which nations emphasize high tech products in their export mix. Figures 6 and 14 profile the 33 countries on TE. TE is the only indicator based solely on statistical components ratio of high tech to total exports and of electronics to total exports. As ratios, both components are bounded so that extremes are not a concern.

Singapore continues to set the standard, with its extreme emphasis on electronics. Malaysia follows, but the remaining countries show more diversification in their exports. The Philippines and Hungary, and to a lesser degree, China, have substantially increased emphasis on high tech exports from 1993 to 1999. Indeed, The Philippines show the highest concentration of high tech in their 1997 export mix of any of the 33 countries.

Interestingly, a number of the high tech powerhouses show somewhat reduced concentration on high tech and electronics in their export mix (e.g., U.S., Japan, Germany, UK, France, Switzerland, South Korea).

Rate of Technological Change (RTC) is intended to be a rate of change measure. As such, countries beginning with lower Technological Standing can more easily show high rates of change. Not surprisingly, this is, by far, the least stable indicator. RTC is composed of three components change in high tech exports, change in electronics exports, and change in expert opinion on its overall technology-intensive production. The two statistical components are the more volatile, even though these reflect 3-period smoothed values (see Appendix for indicator details). S-scores on change in high-tech exports correlate strongly with change in electronics exports (0.66), but each correlates weakly and negatively with the expert opinion change (-0.19 for high tech and 0.12 for electronics).

Figure 7 shows Hungary as the current RTC leader (very high on both statistical measures). Figure 15 shows how volatile this measure is. In 1993, Indonesia was the runaway leader, followed by China. In 1996, Mexico spurted, with Indonesia, Australia, and Russia booming too. For 1999, we note a tempering in RTC from 1996 for each of the Asian Tigers and Cubs, and for the Latin American nations included.

Present vs. 15-Years from Now: As in previous years, we asked our expert panel to estimate high tech production capability at present and in 15 years for each of eight sectors and overall. Figure 8 compares present and 15-year overall estimates provided by the 1999 panel. Keep in mind that these are simply subjective judgments (responses on 1-5 scales, averaged and multiplied by 10 to yield a maximum score of 50). As such there is a marked ceiling effect. For instance, Japan, currently gauged at about 47 can at most be projected to increase 3 points. Nonetheless, the message conveyed here is striking excepting Germany and the UK, every country is expected to increase its high tech export capability over the next 15 years. For most of these 33 nations, the projected gains are large e.g., for all groups except the Big Three and the highly industrialized Western European countries. The anticipated result over the coming decades is a marked broadening of the high tech playing field.

Country Set Comparisons The previous sections discussed results for each indicator. Here we seek to gain perspective by considering sets of countries, across indicators for 1993,1996,1999.

The U.S. has fared extremely well during the 1990’s according to both our input and output indicators. TE and RTC show declines, but these do not pose significant reasons for concern. TE shows similar broadening for the other leading technological exporters as well. This may hint at the emergence of service exports in the global economy, but further research is needed. As mentioned, RTC is both highly volatile and biased against countries with large high tech export bases.

HTI were developed to track the emergence of industrializing nations; comparisons among the existing leading nations are only secondary. We do not include all the leading OECD nations. Having said that, we note general stability in HTI for the leading nations (The Big Three and Western Europe the first two groupings in each chart). TI and PC differentiate within this group far more than do NO and SE with the U.S. and Japan notably outdistancing the others. The American and Japanese dominance is even more striking on TS technological standing (Figure 13) and its key component, high tech exports.

The third grouping in the charts consists of four "English heritage" nations Canada, Australia, South Africa, and New Zealand. All except South Africa show marked increases in NO, very high SE, solid TI, and significant PC. In other words, these are potential high tech competitors of the future. At present, TS (Figure 13) and RTC (Figure 15) show Canada making great strides (it is the leading gainer on the survey item on high tech capability from 1996 to 1999). [As with all these results, obviously our indicators cannot tell the whole story that requires extensive knowledge of manifold factors within a country.]

The fourth grouping is made up of four Eastern European countries Russia, Poland, Hungary, and the Czech Republic. They too display national orientation to compete in high tech (NO), with improving SE, TI, and PC (excepting some decline in TI for Russia). Their future high tech prospects appear bright, even though present TS is weak.

The fifth group consists of the Three Tigers Singapore, South Korea, and Taiwan. Look at their profile across Figures 9-15. Their NO is pronounced; SE and PC are top tier; TI trails the leading technological countries. They have "arrived" Singapore trails only four countries on TS; South Korea and Taiwan evidence strength comparable to most Western European nations. (Our electronics emphasis favors them.) Their TE is pronounced (Singapore leads all).

The sixth group includes six Asian economies of considerable diversity Malaysia, China, Thailand, Indonesia, The Philippines, and India. Again, a scan across the HTI generally shows strong NO, lagging SE, and lagging TI and PC (but note strong advances by China and India). China’s EDP purchases rank fifth in the 33-country set and its electronics production ranks third. China’s 1999 jump on TS is remarkable, prodded particularly by its electronics exports (also third in our set of countries). TE is uneven for these industrializing nations, but RTC has shown strong growth in 1993 and 1996, less so for 1999. The Asian Cubs are growing, albeit unevenly, in high tech competitiveness.

HTI do not show general "Asian Contagion" effects for either the Asian Tigers or Cubs. Output indicators (particularly RTC) show some likely effects on some component measures. Input indicators do not, implying that the Asian nations are not backing off commitment to build potent high tech futures.

Latin America, our seventh group, generally lags on the input and output indicators.

Two newcomers to the HTI, Ireland (part of the Western European group) and Israel, show significant technological competitiveness as smaller economies. They stand forth as highly committed to high tech (NO) and quite strong on the other input indicators. Present capabilities (TS) are also notable (Figure 13).

Concluding Observations

The HTI show reassuring consistency across time, excepting the volatile RTC measure.

The U.S. is doing well. In terms of the four input measures and Technological Standing, we have outdistanced our nearest competitors from 1993 to 1999 according to these measures.

When the HTI development was initiated in the mid-1980’s, a small clique of technologically advanced nations dominated. The sense in profiling a country set including newly industrializing countries was of a "ski slope." High tech exporting "belonged to" the leading OECD countries. The present results might be likened to a gentler "beginners’ ski slope" competition is real (e.g., Malaysia exports more high tech than Italy). [Again, we acknowledge that interpretation is not straightforward; much of Malaysia’s exports come from multinational companies headquartered elsewhere; however the data show that the country has moved well beyond the manufacturing platform model of some years ago.]

The projections of dramatically broadening, international high tech competition convey critical implications for corporate planning and government policy making. The profiles vary by sector but the overall pattern is compelling: National high tech competition is shifting from a steep slope to a broad plateau. High tech production will diffuse dramatically over the coming years. No longer will a few leading nations dominate (to pursue the metaphor -- no skiing in 2015). Japan and the US are projected to remain in the lead, but the gap will close across the board as nations continue to invest in the factors that enhance their ability to compete in high tech products internationally.

Comparison with the Council on Competitiveness Innovation Index. In 1999 the Council on Competitiveness published The New Challenge to America’s Prosperity: Findings from the Innovation Index, by Michael E. Porter and Scott Stern (Washington, D.C.). Comparison with HTI is intriguing. The "innovation index" is constructed by weighting 8 compiled statistical measures (e.g., R&D funding, openness to trade, GDP per capita p. 79) based on their relationship to international patenting activity. It provides "an indication of the relative capability of the economy to produce innovative outputs." The index is constructed with historical data from 1973 for 17 OECD nations and calculated for 8 emerging economies with more limited data since 1990. Thus, their model is best suited for highly developed economies; less so for emerging economies. The HTI model was developed explicitly to forecast technological competitiveness for emerging economies; it is secondarily of interest for OECD countries.

Their conceptual model shares our tenet that "input indicators" can provide leading indicators of future technologically based competitiveness but their regression model links input indicators to 3-year lagged innovative outputs, vs. our roughly 15-year horizon. Their input indicators concentrate very heavily on R&D, whereas ours treat four varied facets of developing technological competitiveness.

We would treat their target output indicator, patenting, as an input indicator, with our output measures keying on technology-based export competitiveness. Emphasis on patenting greatly downplays the competitiveness of potent emerging economies such as China. Their index is normalized (per capita measures), whereas ours is not (most of our statistical components reflect national totals). HTI address national technological competitiveness without particular concern for an economy’s size. In contrast, their innovative capacity is per capita.

These distinctions duly noted, it is fascinating to compare HTI with the "Innovation Index." Table 1 tabulates HTI values for 1993, 1996, and 1999. It then lists the Innovation Index values for 1995 along with change pointers toward 2005.² Some observations:

Next Steps. At this time, with NSF support, we are initiating a review and revision of the HTI. Considerations include: